1. Field of the Invention
The present invention relates to a technique for transmitting digital data.
2. Description of the Related Art
In a digital cable communication system, a binary transmission using the Time Division Multiplex (TDM) method has been conventionally the mainstream in which a large capacity transmission has been realized by a parallel transmission or a high-speed transmission. When the parallel transmission faces a physical limit, a serial transmission, i.e., a high-speed transmission at a data rate of several to more than 10 Gbps by using a high-speed interface (I/F) circuit is performed. However, there is also a limit to the increase in the speed of the data rate, causing a problem of high-frequency loss of a transmission line and deterioration of BER (Bit Error Rate) by reflection.
On the other hand, in a digital RF (Radio Frequency) communication system, multi-bit information is transmitted/received by being embedded in a carrier signal. That is, the data rate is not directly restricted by a carrier frequency. For example, in the QAM (Quadrature Amplitude Modulation) transmission method, most basic quadrature modulation/demodulation method, a 4-level transmission can be realized by a single channel. In 64-QAM, a 64-level transmission can be realized by one carrier signal. That is, a transmission capacity can be improved by such a multi-level modulation method without increasing the carrier frequency.
Such a modulation/demodulation method can be used in a cable communication system without being limited to a radio communication system, the use of which has already been started as PAM (Pulse Amplitude Modulation) system, QPSK (Quadrature Phase Shift Keying) system or DQPSK (Differential QPSK) system. In particular, in an optical communication field, it is important how much information can be embedded in a single optical fiber in terms of costs, and hence, a technical trend is shifting from the binary TDM to a transmission using these digital modulation methods. There is a possibility that, in the near future, these digital multi-level modulation/demodulation methods may be used in cable interfaces between devices including a memory and a SoC (System on a Tip).
[Patent Document 1] U.S. Pat. No. 5,652,552
Because a conventional quadrature amplitude modulator is required to be structured by using a high-speed device, design of the modulator is not easy, or a high-frequency bipolar process or Bi-CMOS process is needed. Therefore, there is a problem that production cost of the device becomes high.